The Composer is not the Expert on Their Own Piece

THE COMPOSER IS NOT THE EXPERT ON THEIR OWN PIECE

Sound Morphology

I want to state something plainly that the last essay in this series only circled.

The person who wrote the piece is not the leading authority on it. On some questions they are the worst-positioned person in the building. And the field is organized almost entirely around pretending otherwise.

This is not a claim about intention or meaning, which are old arguments and largely settled ones. It is narrower and more uncomfortable than that. It is a claim about facts. There are specific, checkable, empirical questions about a piece of music where the composer's answer is systematically less reliable than someone else's, and we go on asking the composer anyway, because the alternative would require building institutions we have never bothered to build.

WHAT THE COMPOSER CANNOT KNOW

Start with difficulty, since that is where the previous essay left off.

Climbers grade routes, and no climbing community on earth accepts the first ascensionist's grade as final. Everyone who subsequently attempts the route votes, and the number drifts toward what the accumulated, embodied experience of many bodies actually reports. The reasoning is not complicated. The person who established the route solved it with their particular reach, their particular finger strength, their particular tolerance for exposure, over however many sessions it took, and by the time they were done they could no longer perceive what was hard about it. Their proximity is precisely what disqualifies them.

A composer is in exactly this position and has none of the correction.

I have spent, in a given piece, four months with a passage. I know which note came first and which arrived on the ninth revision. I know the fingering I imagined, which is almost always the fingering nobody uses. I know what the bar is for, structurally, and that knowledge silently smooths the difficulty every time I look at it, because I am not reading the notation, I am recognizing it. A performer opening that page has none of that scaffolding. They have ink, a deadline, and a body. Their encounter with the material is the real one. Mine was over before the piece existed.

So when I say a passage is difficult, or worse, when I say it is not that difficult, I am reporting on a mental object that has ceased to resemble what is on the page. And I will keep doing this for as long as anyone keeps asking me, because I am not lying. I simply cannot see it anymore.

THE LIST OF QUESTIONS

It is worth being specific about where composer authority actually fails, because the claim is not that we know nothing about our own work.

Where the difficulty is. Not whether the piece is hard, which any competent reader can see, but where the crux sits. Almost every composer I know has been surprised by this at least once. You brace for the passage you agonized over, and the ensemble sails through it and stops dead at three bars you wrote in an afternoon and never thought about again. Fifty performers hitting the same unexpected wall is data. One composer's expectation is a guess.

How long it takes to learn. Composers estimate rehearsal requirements badly and almost always in the same direction. We have no feedback loop that would correct this, because we are rarely in the room for the hours that matter, which are the private ones.

What is actually audible. This is the most uncomfortable item on the list. The composer knows what is written. That knowledge makes it very hard to assess what a listener without the score receives. The thirteenth-order tuplet is on the page. Whether it arrives as a distinguishable rhythmic identity or as a texture is an empirical question, and the person who wrote the ratio is the last person who can answer it, because they will always hear the intention through the result.

What the notation communicates. Not what it means. What it transmits. If eight performers independently read a marking the same wrong way, the marking is wrong. The composer's intent is not a defense. It is not even relevant. A specification that fails to specify has failed, and the only people who can detect the failure are the people who were not present when it was written.

Whether the piece works. Composers are famously unreliable here in both directions, championing the pieces that cost the most and dismissing the ones that came easily, as though difficulty of manufacture were a measure of value. It is a measure of manufacture.

WHAT THE COMPOSER IS THE EXPERT ON

The argument would be dishonest without this section.

The composer knows what was rejected, which is often the most valuable unrecorded information about any piece. They know why the fourth movement is short. They know which passages are load-bearing and which are transitional, and therefore where an interpreter has latitude and where they do not. They know what the piece is in conversation with. They know, usually, what would constitute a bad performance, even when they cannot say what would constitute a good one.

That is real expertise and it matters. But look at the shape of it. Every item on that list concerns the piece's history and its architecture. Not one of them concerns its reception, its difficulty, its legibility, or its effect. The composer is the leading authority on how the piece came to exist. They are an ordinary witness, and sometimes a poor one, on what it does now that it does exist.

We routinely conflate those two authorities, and the conflation is not innocent. It is how the first one gets used to settle arguments that belong to the second.

WHY THE FIELD IS BUILT THIS WAY

Because the alternative requires infrastructure, and infrastructure requires admitting that the individual authorial voice is insufficient.

A climbing gym employs route setters who watch. If nobody can do the crux at the intended grade, or if everyone finds it trivial, the route gets reset. The hold angle changes. The foot placement moves. This is a profession built on the premise that a difficulty judgment is a hypothesis tested against real bodies and revised when the bodies disagree.

Music has no equivalent and does not want one. A score is published and its difficulty becomes permanent, along with its reputation, its performance practice, and its errors. There is no mechanism by which fifty performances of the same passage produce a revision. There is no forum in which a claim about a piece can be collectively re-litigated. Some composers revise editions after performer feedback, and when they do it is described as generosity, which tells you everything. In climbing it would be called doing the job.

The result is that we have built an entire critical and pedagogical apparatus on a single unaudited source, and then we have made deference to that source a marker of seriousness. Ask the composer. Consult the composer's writings. Check what the composer said in the interview. All of it treating one person's account as evidence, when in most other domains that account would be considered the least reliable input available, precisely because of the informant's involvement.

THE PERFORMERS ALREADY KNOW THIS

Here is what makes the whole arrangement slightly absurd.

The distributed knowledge exists. It is simply unrecorded. Performers who specialize in difficult repertoire know exactly which pieces are sandbagged, which reputations are inherited rather than earned, which markings everyone quietly ignores, which passages have a standard solution that no score mentions, and which composers write beautifully for the instrument versus which ones write beautifully at it. They know all of this in detail and they mostly say it to each other, in rehearsal breaks, off the record, because saying it publicly reads as complaint, and complaint is a career problem.

So the most accurate available account of what a piece actually is lives in an oral tradition that is deliberately kept out of print, while the least accurate account, the composer's, is published, cited, taught, and consulted.

WHAT I AM ACTUALLY ASKING FOR

Not less authority for composers. A correct allocation of it.

Ask the composer what was rejected, what is structural, what the piece is arguing with. Those are genuinely composer questions.

Do not ask the composer how hard it is, how long it takes, what a listener hears, or whether the notation is clear. Ask the fiftieth performer, and then ask the fifty-first, and treat the aggregate as more authoritative than any single account including the author's. That is not disrespect. It is the ordinary epistemics of every field that takes its own claims seriously.

And write it down. The reason climbing grades converge on something usable is not that climbers are more honest than musicians. It is that they built a public record and argue in it constantly. We have the arguments. We simply hold them in corridors.

I have written pages I was certain were the hardest thing I had ever asked of anyone, and watched them turn out to be nothing. I have written bars I thought were a rest stop and watched three excellent players fail at them in exactly the same place, which is the clearest signal in the world that the fault is mine and located precisely there. Every time, the information came from somebody else. It could not have come from me.

The score is the only thing I actually know. Everything after it belongs to other people, and they are better witnesses than I am.